Re: [Geotools-devel] Mif Mid reader up on CVS.

[<ra...@ei...>]

> The CTS module is stable at this time (until peoples ask me to redesign 
> it if they feel that some stuff should be changed). There was no 
> incompatible change in the API for months and the only work done lately 
> was optimisation and bug fixes. Results are raisonable (no obviously 
> wrong transformations), however more tests are needed in order to check 
> if they are really accurate. There is some doubt about how we should 
> interpret the OGC specs when more than one transformation are chained 
> together.

Great.

> How coordinate systems are coded in MIF file?
> 
The coordinate system in a MIF file is laid out in the header. All 
point information after that is stored as that particular coordinate 
system : 

(from wotsit.org...)


CoordSys Clause
Specify the COORDSYS clause to note that the data is not stored 
in
longitude/latitude form.   When no COORDSYS clause is specified, 
data is
assumed to be stored in longitude/latitude forms. All coordinates 
are stored
with respect to the northeast quadrant. The coordinates for points 
in the
United States have a negative X while coordinates for points in 
Europe (east
of Greenwich) have a positive X coordinates for points in the 
Northern
hemisphere have a positive Y while coordinates for points in the 
Southern
hemisphere have a negative Y.
	Syntax 1:
           CoordSys Earth
            [ Projection type,
                datum,
                unitname
                [,origin_longtitude  ]
                [,origin_lattitude   ]
                [,standard_parallel_1  [,standard_parallel_2]]
                [,azimuth ]
                [,scale_factor  ]
                [,false_easting ]
                [,false_northing]
                [,range  ]]
            [ Bounds (minx,miny) (maxx,maxy) ]
     Syntax 2:
        CoordSys Nonearth
            Units unitname
            Bounds (minx,miny) (maxx,maxy)

thanks,

Ray.